v0.3.5.2

README.md

@@ -5,13 +5,13 @@ Another cross-platform, efficient, practical and pretty CSV/TSV toolkit
 Yes, you could just use spreadsheet softwares like MS excel to
 do most of the job.
 
-Howerver it's all by clicking and typing, which is **not
-automatically and time-consuming to repeate**, especially when we want to
+However it's all by clicking and typing, which is **not
+automatically and time-consuming to repeat**, especially when we want to
 apply similar operations with different datasets or purposes.
 
 `csvtk` is **convenient for rapid investigation
 and also easy to integrated into analysis pipelines**.
- It could save you much time of writting scripts.
+ It could save you much time of writing scripts.
 
 Hope it be helpful to you.
 
@@ -92,7 +92,7 @@ You can add the directory of the executable file to environment variable
 For Go developer, just one command:
 
     go get -u github.com/shenwei356/csvtk/csvtk
-    
+
 ## Compared to `csvkit`
 
 [csvkit](http://csvkit.readthedocs.org/en/540/)
@@ -114,7 +114,7 @@ Similar tools:
 
 - [csvkit](http://csvkit.readthedocs.org/en/540/) - A suite of utilities for converting to and working with CSV, the king of tabular file formats. http://csvkit.rtfd.org/
 - [xsv](https://github.com/BurntSushi/xsv) - A fast CSV toolkit written in Rust.
-- [miller](https://github.com/johnkerl/miller) - Miller is like sed, awk, cut, join, and sort for 
+- [miller](https://github.com/johnkerl/miller) - Miller is like sed, awk, cut, join, and sort for
 name-indexed data such as CSV and tabular JSON http://johnkerl.org/miller
 - [tsv-utils-dlang](https://github.com/eBay/tsv-utils-dlang) - Command line utilities for tab-separated value files written in the D programming language.
 
@@ -157,7 +157,7 @@ Examples
         1         4      20     1       8        5     3.16
         2         4   1,015     2   1,000   253.75   497.51
         3         4      13     0       6     3.25      2.5
-        
+
 1. Select fields/columns (`cut`)
 
     - By index: `csvtk cut -f 1,2`

csvtk/cmd/helper.go

@@ -34,7 +34,7 @@ import (
 )
 
 // VERSION of csvtk
-var VERSION = "0.3.5.1"
+const VERSION = "0.3.5.2"
 
 func checkError(err error) {
 	if err != nil {
@@ -297,15 +297,15 @@ func parseFields(cmd *cobra.Command,
 				negativeFields = true
 			} else {
 				if negativeFields {
-					checkError(fmt.Errorf(`filed should not fixed with positive and negative fields`))
+					checkError(fmt.Errorf(`filed should not mixed with positive and negative fields`))
 				}
 			}
 		}
 		// 2 pass check
 		if negativeFields {
 			for _, f := range fields {
 				if f > 0 {
-					checkError(fmt.Errorf(`filed should not fixed with positive and negative fields`))
+					checkError(fmt.Errorf(`filed should not mixed with positive and negative fields`))
 				}
 			}
 		}

csvtk/cmd/join.go

@@ -62,6 +62,7 @@ Multiple keys supported, but the orders are ignored.
 
 		fuzzyFields := getFlagBool(cmd, "fuzzy-fields")
 		keepUnmatched := getFlagBool(cmd, "keep-unmatched")
+		fill := getFlagString(cmd, "fill")
 
 		outfh, err := xopen.Wopen(config.OutFile)
 		checkError(err)
@@ -137,7 +138,7 @@ Multiple keys supported, but the orders are ignored.
 				} else {
 					if keepUnmatched {
 						for i := 1; i <= len(data[0])-len(fieldsMap); i++ {
-							record = append(record, "")
+							record = append(record, fill)
 						}
 						Data2 = append(Data2, record)
 					}
@@ -165,4 +166,5 @@ func init() {
 	joinCmd.Flags().BoolP("ignore-case", "i", false, `ignore case`)
 	joinCmd.Flags().BoolP("fuzzy-fields", "F", false, `using fuzzy fileds, e.g. *name or id123*`)
 	joinCmd.Flags().BoolP("keep-unmatched", "k", false, `keep unmatched data of the first file`)
+	joinCmd.Flags().StringP("fill", "", "", `fill content for unmatched data`)
 }

csvtk/cmd/stat.go

@@ -26,8 +26,8 @@ import (
 	"github.com/brentp/xopen"
 	"github.com/spf13/cobra"
 
-	"github.com/tatsushid/go-prettytable"
 	"github.com/dustin/go-humanize"
+	"github.com/tatsushid/go-prettytable"
 )
 
 // statCmd represents the stat command
@@ -46,10 +46,9 @@ var statCmd = &cobra.Command{
 		checkError(err)
 		defer outfh.Close()
 
-
 		tbl, err := prettytable.NewTable([]prettytable.Column{
-		    {Header: "file"},
-		    {Header: "num_cols", AlignRight: true},
+			{Header: "file"},
+			{Header: "num_cols", AlignRight: true},
 			{Header: "num_rows", AlignRight: true}}...)
 		checkError(err)
 		tbl.Separator = "   "
@@ -73,9 +72,12 @@ var statCmd = &cobra.Command{
 					once = false
 				}
 			}
-			if !config.NoHeaderRow {
+			if numRows > 0 && !config.NoHeaderRow {
 				numRows--
 			}
+			if numRows < 0 {
+				numRows = 0
+			}
 			tbl.AddRow(
 				file,
 				humanize.Comma(int64(numCols)),

doc/docs/download.md

@@ -6,8 +6,9 @@
 
 ## Current Version
 
-- [csvtk v0.3.5.1](https://github.com/shenwei356/csvtk/releases/tag/v0.3.5.1)
-    - fix minor bug of reading lines ending with `\r\n` from a dependency package
+- [csvtk v0.3.5.2](https://github.com/shenwei356/csvtk/releases/tag/v0.3.6)
+    - add flag `--fill` for `csvtk join`, so we can fill the unmatched data
+    - fix typo
 
 ## Installation
 
@@ -38,6 +39,8 @@ For Go developer, just one command:
 
 ## Previous Versions
 
+- [csvtk v0.3.5.1](https://github.com/shenwei356/csvtk/releases/tag/v0.3.5.1)
+    - fix minor bug of reading lines ending with `\r\n` from a dependency package
 - [csvtk v0.3.5](https://github.com/shenwei356/csvtk/releases/tag/v0.3.5)
     - fix minor bug of `csv2md`
     - add subcommand `version` which could check for update

doc/docs/tutorial.md

@@ -2,18 +2,20 @@
 
 ## Forewords
 
+Another cross-platform, efficient, practical and pretty CSV/TSV toolkit
+
 Yes, you could just use spreadsheet softwares like MS excel to
 do most of the job.
 
-Howerver it's all by clicking and typing, which is **not
-automatically and time-consuming to repeate**, especially when we want to
+However it's all by clicking and typing, which is **not
+automatically and time-consuming to repeat**, especially when we want to
 apply similar operations with different datasets or purposes.
 
 `csvtk` is **convenient for rapid investigation
 and also easy to integrated into analysis pipelines**.
- It could save you much time of writting scripts.
+ It could save you much time of writing scripts.
 
-Hope it be helpful for you.
+Hope it be helpful to you.
 
 ## Analyzing OTU table
 
@@ -107,7 +109,7 @@ What's a mess. Let's make it prettier!
         B.2        .13              .55          .16             .01               .03
         B.3        .22              .41          .29             .01               .02
 
-1. Rename first column
+1. Rename name of the first column
 
         $ csvtk rename -f 1 -n "sample" otu_table.gAB.t.csv -o otu_table.gAB.t.r.csv
 

doc/docs/usage.md

@@ -19,7 +19,7 @@ Usage
 ```
 Another cross-platform, efficient and practical CSV/TSV toolkit
 
-Version: 0.3.5
+Version: 0.3.5.2
 
 Author: Wei Shen <[email protected]>
 
@@ -473,6 +473,7 @@ Usage:
 
 Flags:
   -f, --fields string    Semicolon seperated key fields of all files, if given one, we think all the files have the same key columns. e.g -f 1;2 or -f A,B;C,D or -f id (default "1")
+      --fill string      fill content for unmatched data
   -F, --fuzzy-fields     using fuzzy fileds, e.g. *name or id123*
   -i, --ignore-case      ignore case
   -k, --keep-unmatched   keep unmatched data of the first file